Exploring parameter interactions and deep learning for modeling pressure related downhole safety conditions during drilling
Bibliographic record
Abstract
Drilling for petroleum is technically engaging, considering the potentially huge risk involved. Blowout due to uncontrolled kick represents a scenario that is to be avoided due to disastrous consequences, e.g. huge financial loss, environmental damage, and death of personnel. Kick occurrence can be prevented if the pore pressure is correctly estimated and the proper drilling mud weight employed. Pore pressure prediction is done in shale lithology; hence a fast means of proper lithology identification is important for pore pressure prediction. Monitoring downhole for pore pressure related hazard therefore includes but is not limited to: monitoring for kick occurrence, monitoring for abnormal pore pressure, and monitoring for changes in lithology for adequate pore pressure prediction. In the field of data science, deep learning is gaining significant interest, which is likely due to its potentials and successful applications. Researchers have begun to explore deep learning in several areas with close affinity to drilling engineering, such as lithology identification, drilling rig state determination, generating logging/other drilling parameters, detecting downhole events, and detecting abnormality in data. Therefore, this serves as a motivation to take advantage of deep learning capabilities in monitoring downhole conditions during drilling to prevent pore pressure based hazardous events. In this dissertation, a novel methodology for kick detection using drilling parameters is presented. Likewise, a novel methodology for predicting the shaliness of a rock formation using drilling parameters is also presented. These methodologies utilized deep learning algorithms in order to achieve the desired objectives. Results obtained using field data justified the development of methodologies with the capability to capture sequential dependencies. Cost represents a significant factor for utilizing drilling parameters in comparison to the use of highly sophisticated/expensive downhole sensors. As part of this dissertation, a novel approach for pore pressure prediction from porosity and resistivity measurement is presented. The aim of combining porosity and resistivity is to explore how the interrelationship between them can enhance pore pressure prediction. The methodology developed for combining porosity and resistivity performed better than the conventional approach based on field data. Machine learning was also employed for pore pressure prediction and better result was also achieved in comparison to conventional approach based on the same field data. In summary, this dissertation presents several novel methodologies for monitoring different aspects of downhole conditions from downhole lithology to downhole drilling events which are important for improved drilling safety.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".